June 2020
Volume 61, Issue 7
Open Access
ARVO Annual Meeting Abstract  |   June 2020
Kalman Filtering Based Machine Learning to Predict Future Mean Deviation Values for Patients with Glaucoma -- Enhancing Existing Models Using Data from Optical Coherence Tomography. A Study Using Data from ADAGES and DIGS.
Author Affiliations & Notes
  • Mohammad Zhalechian
    Industrial Engineering, University of Michigan, Ann Arbor, Michigan, United States
  • Mark P Van Oyen
    Industrial Engineering, University of Michigan, Ann Arbor, Michigan, United States
  • Mariel S Lavieri
    Industrial Engineering, University of Michigan, Ann Arbor, Michigan, United States
  • Carlos Gustavo De Moraes
    Bernard and Shirlee Brown Glaucoma Research Laboratory, Columbia University, New York, United States
  • Christopher A Girkin
    Dept. of Ophthalmology, University of Alabama-Birmingham, Alabama, United States
  • Massimo Antonio Fazio
    Dept. of Ophthalmology, University of Alabama-Birmingham, Alabama, United States
  • Robert N Weinreb
    The Viterbi Family Department of Ophthalmology, Univ of California-San Diego, California, United States
    The Shiley Eye Institute, University of California - San Diego, California, United States
  • Christopher Bowd
    The Viterbi Family Department of Ophthalmology, Univ of California-San Diego, California, United States
    The Shiley Eye Institute, University of California - San Diego, California, United States
  • Jeffrey M Liebmann
    Ophthalmology, Columbia University Irving Medical Center, New York, United States
  • Linda M Zangwill
    The Viterbi Family Department of Ophthalmology, Univ of California-San Diego, California, United States
    The Shiley Eye Institute, University of California - San Diego, California, United States
  • Chris A Andrews
    Medical School, University of Michigan, Michigan, United States
    Center for Eye Policy and Innovation, University of Michigan, Michigan, United States
  • Joshua D Stein
    Medical School, University of Michigan, Michigan, United States
    Center for Eye Policy and Innovation, University of Michigan, Michigan, United States
  • Footnotes
    Commercial Relationships   Mohammad Zhalechian, None; Mark Van Oyen, National Eye Inst (F); Mariel Lavieri, National Eye Inst (F); Carlos De Moraes, Belite (C), Carl Zeiss (C), Galimedix (C), Heidelberg (R), Novartis (C), Perfuse Therapeutics (C), Reichert (C), Topcon (R); Christopher Girkin, EyeSight Foundation of Alabama (F), Heidelberg Engineering, GmbH (F), National Eye Institute (F), Research to Prevent Blindness (F); Massimo Fazio, EyeSight Foundation of Alabama (F), GmbH (F), Heidelberg Engineering (F), National Eye Institute (F), Research to Prevent Blindness (F); Robert Weinreb, Aerie Pharmaceuticals (C), Allergan (C), Bausch&Lomb (F), Bausch&Lomb,Eyenovia (C), Carl Zeiss Meditec (F), Centervue (F), Eynovia (C), Heidelberg Engineering (F), Konan Medical (F), Meditec-Zeiss (P), Optovue (F), Toromedes (P); Christopher Bowd, None; Jeffrey Liebmann, Aerie (C), Alcon, Inc. (C), Allergen (C), Bausch & Lomb (F), Bausch & Lomb, Inc. (C), Carl Zeiss Meditec (F), Carl Zeiss Meditech, Inc. (C), Eyenova (E), Galimedex (C), Heidelberg Engineering (F), Heidelberg Engineering GmbH (C), National Eye Institute (F), Novartis (C), Optovue (F), Reichert (C), Reichert (F), Research to Prevent Blindness (F), Topcon (F); Linda Zangwill, Carl Zeiss Meditec Inc. (F), Heidelberg Engineering (R), Heidelberg Engineering GmbH (F), Meditec-Zeiss (P), National Eye Institute (F), Optovue Inc. (F), Topcon Medical Systems Inc. (F); Chris Andrews, National Eye Inst (F); Joshua Stein, National Eye Inst (F)
  • Footnotes
    Support  R01EY026641, EY11008, EY19869, EY14267, EY027510, EY026574, EY029058, P30EY022589, Unrestricted grant from Research to Prevent Blindness, New York, New York
Investigative Ophthalmology & Visual Science June 2020, Vol.61, 1986. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Mohammad Zhalechian, Mark P Van Oyen, Mariel S Lavieri, Carlos Gustavo De Moraes, Christopher A Girkin, Massimo Antonio Fazio, Robert N Weinreb, Christopher Bowd, Jeffrey M Liebmann, Linda M Zangwill, Chris A Andrews, Joshua D Stein; Kalman Filtering Based Machine Learning to Predict Future Mean Deviation Values for Patients with Glaucoma -- Enhancing Existing Models Using Data from Optical Coherence Tomography. A Study Using Data from ADAGES and DIGS.. Invest. Ophthalmol. Vis. Sci. 2020;61(7):1986.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose : To assess whether it is possible to improve the accuracy of predicting future values of mean deviation (MD) for glaucoma suspects and patients with open-angle glaucoma (OAG) by enhancing our previously developed machine learning based Kalman filtering (ML-KF) approach with data on retinal nerve fiber layer (rNFL) thickness from optical coherence tomography (OCT).

Methods : We identified 109 glaucoma suspects and 438 patients with OAG enrolled in ADAGES/DIGS who received tonometry, perimetry, and rNFL OCT approximately every 6 months for 6 years. We parameterized, calibrated, and validated 2 ML-KF models to predict MD on standard automated perimetry 3 years into the future and compared them to actual MD values obtained at study visits. Our previously created ML-KF (KF_A) considered only past values of MD, PSD, and IOP from the patient while a new ML-KF (KF_A+OCT) used these inputs plus past rNFL thickness measurements. Both were compared against 2 traditional linear regression (LR) forecasting models. We compared the error distribution and RSME of the 4 models to determine which was most accurate.

Results : When tested on the 438 patients with OAG, the proportions of MD errors within 1.0 dB of the actual value at 3 years into the future for KF_A, KF_A+OCT, LR1, and LR2 were 43%, 46%, 31%, and 30%, respectively, and the RMSEs were 3.87, 3.88, 4.46, and 4.34, respectively. When the same comparisons were made on the 109 glaucoma suspects, results were 70%, 76%, 52%, and 52% for KF_A, KF_A+OCT, LR1, and LR2, and the RMSEs were 1.05, 0.94, 1.50, and 1.50, respectively.

Conclusions : Our ML-KF models are capable of predicting values of MD 3 years into the future to within 1.0 dB of the actual value for nearly half of the ADAGES/DIGS study participants and within 2.5 dB of the actual value for 83%. The addition of rNFL data from OCT into our ML-KF models enhanced their predictive ability for glaucoma suspects more than for patients with OAG.

This is a 2020 ARVO Annual Meeting abstract.

×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×